Power generation group industrial control system supply chain safety management method, system, equipment and medium

Abstract

The invention discloses a power generation group industrial control system supply chain safety management method, system and device and a medium. The method comprises the following steps: S101, obtaining supply chain related data of a power generation group industrial control system; s102, anomaly detection is conducted on the supply chain related data through an auto-encoder model, the auto-encoder model is a pruned auto-encoder model, and the method, the system, the equipment and the medium can solve the problem that in the existing equipment purchasing process, supplier safety management is limited.

Description

Technical Field

[0001] The present invention belongs to the field of electric power technology and relates to a supply chain security management method, system, equipment and medium for an industrial control system of a power generation group. Background Art

[0002] With the increasing application of emerging information technologies such as the Industrial Internet and the Internet of Things in the energy field, power grid companies have transformed from traditional closed production and operation models to open distributed production and operation models, which has led to a surge in the number of smart device connections and increasingly complex device types, posing huge challenges to the security of smart devices. At the same time, with the increasing number of smart devices, device suppliers are becoming more and more diverse, and products provided by different manufacturers may have different security risks. Therefore, it is necessary to strengthen the security management of suppliers to ensure the normal operation of power companies.

[0003] At present, in order to strengthen the security management of equipment suppliers, third-party organizations are usually introduced to conduct assessments during the equipment procurement process, and audit reports are used to ensure equipment security. However, this method can only audit a single device and cannot conduct an overall analysis of the entire supply chain, which has certain limitations. Summary of the invention

[0004] The purpose of the present invention is to overcome the shortcomings of the above-mentioned prior art and to provide a method, system, equipment and medium for supply chain security management of an industrial control system of a power generation group. The method, system, equipment and medium can solve the problems of limitations in supplier security management in the existing equipment procurement process.

[0005] To achieve the above object, the present invention discloses a method for supply chain security management of an industrial control system of a power generation group, comprising the following steps:

[0006] S101: Acquire supply chain related data of the industrial control system of the power generation group;

[0007] S102: Perform anomaly detection on the supply chain related data using an autoencoder model, wherein the autoencoder model is a pruned autoencoder model.

[0008] Furthermore, the acquisition process of the autoencoder model is:

[0009] S201: Obtain an initial autoencoder model, wherein the initial autoencoder model includes a plurality of neuron layers, and each neuron layer includes at least one neuron unit;

[0010] S202: for each neuron layer in the initial autoencoder model, determine whether any neuron unit in the neuron layer satisfies the first condition, if not, execute step S203, if yes, execute step S204;

[0011] S203: deleting the neuron unit in the neuron layer;

[0012] S204: Adjusting the parameters corresponding to the neuron unit in the neuron layer.

[0013] Furthermore, the first condition is any one or all of the second condition and the third condition, wherein the second condition is that the first parameter corresponding to the neuron unit is less than a first threshold, and the third condition is that the second parameter corresponding to the neuron unit is greater than a second threshold.

[0014] Furthermore, the adjusting of the parameters corresponding to the neuron unit in the neuron layer includes: when a first parameter corresponding to the neuron unit is less than a first threshold and a second parameter corresponding to the neuron unit is not greater than a second threshold, setting the first parameter corresponding to the neuron unit to zero.

[0015] Furthermore, the initial autoencoder model also includes at least one skip connection, and the skip connection is used to connect two non-continuous neuron layers in the initial autoencoder model.

[0016] Furthermore, the process of anomaly detection is as follows:

[0017] Reconstructing the supply chain related data using the autoencoder model to obtain reconstructed data;

[0018] calculating a distance between the supply chain related data and the reconstructed data as a reconstruction error;

[0019] When the reconstruction error is greater than a reconstruction error threshold, the supply chain related data is determined to be abnormal data.

[0020] Furthermore, the anomaly detection also includes:

[0021] Counting the reconstruction errors of all samples in the sample set to which the supply chain-related data belongs, and the average reconstruction errors of all samples in the sample set;

[0022] When the average reconstruction errors of all samples in the sample set are greater than or equal to a sample average error threshold, the sample set is determined to be an abnormal sample set.

[0023] The present invention discloses a power generation group industrial control system supply chain security management system, including:

[0024] An acquisition module, used to acquire supply chain related data of the industrial control system of the power generation group;

[0025] An anomaly detection module is used to perform anomaly detection on the supply chain related data using an autoencoder model, wherein the autoencoder model is a pruned autoencoder model.

[0026] The present invention discloses a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the supply chain security management method of the industrial control system of the power generation group are implemented.

[0027] The present invention discloses a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the supply chain security management method of the industrial control system of a power generation group are implemented.

[0028] The present invention has the following beneficial effects:

[0029] The supply chain security management method, system, device and medium of the power generation group industrial control system described in the present invention obtains the supply chain related data of the power generation group industrial control system during specific operation; then, the supply chain related data is detected for anomalies using an autoencoder model, wherein the autoencoder model is a pruned autoencoder model. It can be seen that the present invention uses an autoencoder model to perform anomaly detection on supply chain related data, which can quickly discover abnormal situations in the supply chain and improve power supply reliability. At the same time, it can also reduce maintenance costs and improve operation and maintenance efficiency. In addition, the present invention uses a pruned autoencoder model for anomaly detection, which can reduce the amount of data required for model training, shorten the model training time, and improve the model prediction accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The accompanying drawings constituting a part of the present invention are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the accompanying drawings:

[0031] Figure 1 The present invention is a flow chart of the method. DETAILED DESCRIPTION

[0032] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0033] In the description of the present invention, it should be understood that the terms “include” and “comprises” indicate the presence of described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or collections thereof.

[0034] It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0035] It should be further understood that the term "and / or" used in the present specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes these combinations. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in the present invention generally indicates that the associated objects are in an "or" relationship.

[0036] It should be understood that, although the terms first, second, third, etc. may be used to describe preset ranges, etc. in the embodiments of the present invention, these preset ranges should not be limited to these terms. These terms are only used to distinguish preset ranges from each other. For example, without departing from the scope of the embodiments of the present invention, the first preset range may also be referred to as the second preset range, and similarly, the second preset range may also be referred to as the first preset range.

[0037] The word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.

[0038] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention described and shown in the drawings here can usually be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0039] Various structural schematic diagrams of the embodiments disclosed in the present invention are shown in the accompanying drawings. These figures are not drawn to scale, and some details are magnified and some details may be omitted for the purpose of clear expression. The shapes of various regions and layers shown in the figures and the relative sizes and positional relationships therebetween are only exemplary, and may deviate in practice due to manufacturing tolerances or technical limitations, and those skilled in the art may additionally design regions / layers with different shapes, sizes, and relative positions according to actual needs.

[0040] refer to Figure 1 The power generation group industrial control system supply chain security management method of the present invention comprises the following steps:

[0041] S101: Acquire supply chain related data of the industrial control system of the power generation group;

[0042] S102: Perform anomaly detection on the supply chain related data using an autoencoder model, wherein the autoencoder model is a pruned autoencoder model.

[0043] Specifically, in the embodiment of the present application, supply chain related data of the power generation group's industrial control system can be obtained through various channels, for example, it can be obtained from relevant documents issued by superior departments, it can be obtained by directly asking relevant personnel, it can also be obtained from the company's internal database, and so on.

[0044] In practical applications, since the supply chain-related data of the power generation group's industrial control system usually involves a large amount of information, such as equipment name, equipment model, supplier name, supplier address, supplier contact information, supplier qualification certificate, etc., this information may come from different sources and the formats may also be different. Therefore, after obtaining the supply chain-related data, it is necessary to pre-process it for subsequent analysis.

[0045] Specifically, the embodiment of the present application can use an autoencoder model to detect anomalies in the supply chain related data. Specifically, the autoencoder is an unsupervised learning algorithm that learns the intrinsic representation of data by compressing and decompressing the input data. The basic structure of the autoencoder consists of two parts: an encoder and a decoder. The encoder is responsible for compressing the original data into a set of hidden state vectors, and the decoder is responsible for restoring the hidden state vector to the original data. In this process, the autoencoder attempts to minimize the reconstruction loss function, that is, the difference measure between the original data and the reconstructed data. When the error of a sample point after reconstruction is large, it indicates that it is an outlier.

[0046] In order to improve the performance of the autoencoder, the embodiment of the present application adopts a pruned autoencoder model for anomaly detection. Specifically, pruning refers to the operation of removing some redundant nodes (such as convolution kernels, fully connected layers) without affecting the accuracy of the network. Through the pruning operation, not only can the model size be reduced and the reasoning speed be accelerated, but also the occurrence of overfitting can be prevented. In an embodiment of the present application, the prune_low_magnitude() function in the deep learning framework TensorFlow can be used to complete the pruning operation.

[0047] In summary, the embodiment of the present application uses an autoencoder model to perform anomaly detection on supply chain related data, which can quickly discover abnormal situations in the supply chain, improve power supply reliability, and also reduce maintenance costs and improve operation and maintenance efficiency. In addition, the embodiment of the present application uses a pruned autoencoder model for anomaly detection, which can reduce the amount of data required for model training, shorten the model training time, and improve the model prediction accuracy.

[0048] In an optional embodiment, the autoencoder model is obtained as follows:

[0049] S201: Obtain an initial autoencoder model, wherein the initial autoencoder model includes a plurality of neuron layers, and each neuron layer includes at least one neuron unit;

[0050] S202: for each neuron layer in the initial autoencoder model, determine whether any neuron unit in the neuron layer satisfies the first condition, if not, execute step S203, if yes, execute step S204;

[0051] S203: deleting the neuron unit in the neuron layer;

[0052] S204: Adjusting the parameters corresponding to the neuron unit in the neuron layer.

[0053] Specifically, in the embodiment of the present application, an initial autoencoder model can be defined first, and then the model can be pruned. The "pruning" mentioned here refers to removing some unnecessary neuron units, thereby reducing the model size and improving operational efficiency. It should be noted that pruning does not mean arbitrarily deleting neuron units, but ensuring the accuracy of the model to a certain extent. To this end, we can calculate the loss function value of the current model after each iteration, and compare it with the result of the previous iteration. If the loss function value this time is smaller than the last time, the current model is retained, otherwise the model continues to be optimized.

[0054] Specifically, in the embodiment of the present application, an initial autoencoder model can be defined first, which includes multiple neuron layers, and each neuron layer includes several neuron units. Then, each neuron layer is traversed to check whether there are some neuron units with very small weights. If so, you can consider deleting them. This is because neuron units with very small weights often do not have much impact on the final result, but they will increase the computational complexity and memory usage of the model. Therefore, the pruning operation can effectively reduce this waste and improve the efficiency of the model.

[0055] Next, the remaining neuron units need to be adjusted to make them more suitable for the needs of the current task. Specifically, the gradient descent method can be used to update the parameter values ​​of each neuron unit so that they better fit the training data. It is worth noting that overfitting may occur in this process, that is, the model is too close to the training data and has poor generalization ability. To avoid this, a regularization term can be added to constrain the complexity of the model.

[0056] In summary, through the above steps, a pruned autoencoder model is obtained, which can be used to identify anomalies in supply chain related data.

[0057] In an optional embodiment, the first condition is any one or all of the second condition and the third condition, wherein the second condition is that the first parameter corresponding to the neuron unit is less than a first threshold, and the third condition is that the second parameter corresponding to the neuron unit is greater than a second threshold.

[0058] Specifically, in the embodiments of the present application, different pruning strategies can be flexibly selected according to actual conditions. For example, if you want to keep the overall architecture of the model unchanged and simply remove some unnecessary neuron units, you can choose to sort them by weight and only keep the top-ranked ones. Another common approach is to use a sparse initialization method to randomly initialize a batch of neuron units with smaller weights, and then gradually increase their number until the desired effect is achieved.

[0059] In addition, the complexity of the model can be controlled by combining regularization terms. Specifically, assume that there is an L2-norm regularization term λ||w||^2, where w represents the weight matrix of the neuron unit and λ is a hyperparameter. During the training process, the value of w is constantly adjusted to make the loss function J(w) as small as possible while trying to avoid w becoming too large to cause overfitting. Therefore, by reasonably setting the value of λ, the complexity and generalization ability of the model can be balanced.

[0060] In an optional embodiment, the adjusting the parameters corresponding to the neuron unit in the neuron layer includes:

[0061] If the first parameter corresponding to the neuron unit is less than a first threshold and the second parameter corresponding to the neuron unit is not greater than a second threshold, the first parameter corresponding to the neuron unit is set to zero.

[0062] Specifically, in the embodiment of the present application, after pruning, there are still complex interdependencies between the remaining neuronal units. Therefore, fine-tuning is required again so that the model can better fit the training data on the new basis. Generally speaking, fine-tuning is divided into two cases: one is to fix the trained parameters and only update the newly added parameters; the other is to update all parameters at the same time. In practice, we often try the second method first because it allows the model to converge faster and is not easy to fall into the local optimal solution.

[0063] In an optional embodiment, the initial autoencoder model further includes at least one skip connection, and the skip connection is used to connect two non-continuous neuron layers in the initial autoencoder model.

[0064] Specifically, in addition to conventional neuron layers, some special components are sometimes introduced in the embodiments of the present application, such as skip connections, which can directly transfer the output of a previous layer to the position of a certain number of layers behind. This has two advantages: first, it can alleviate the gradient vanishing problem, because the skip connection is equivalent to increasing the number of back propagation paths; second, it can make it easier for the model to learn some long-range dependencies, which is very important for fields such as natural language processing.

[0065] In an optional embodiment, the anomaly detection includes:

[0066] Reconstructing the supply chain related data using the autoencoder model to obtain reconstructed data;

[0067] calculating a distance between the supply chain related data and the reconstructed data as a reconstruction error;

[0068] When the reconstruction error is greater than a reconstruction error threshold, the supply chain related data is determined to be abnormal data.

[0069] Specifically, an autoencoder model can be used in an embodiment of the present application to detect anomalies in supply chain related data. Specifically, an autoencoder is an unsupervised learning algorithm that learns the intrinsic representation of data by compressing and decompressing the input data. The basic structure of an autoencoder consists of two parts: an encoder and a decoder. The encoder is responsible for compressing the original data into a set of hidden state vectors, and the decoder is responsible for restoring this hidden state vector to the original data. In this process, the autoencoder attempts to minimize the reconstruction loss function, which is a measure of the difference between the original data and the reconstructed data. If the error of a sample point after reconstruction is large, it indicates that it is an outlier.

[0070] In an optional embodiment, the anomaly detection further includes:

[0071] Counting the reconstruction errors of all samples in the sample set to which the supply chain-related data belongs, and the average reconstruction errors of all samples in the sample set;

[0072] When the average reconstruction errors of all samples in the sample set are greater than or equal to a sample average error threshold, the sample set is determined to be an abnormal sample set.

[0073] Specifically, in the embodiments of the present application, an autoencoder model can be used to detect anomalies in supply chain related data. Specifically, an autoencoder is an unsupervised learning algorithm that learns the intrinsic representation of data by compressing and decompressing the input data. The basic structure of an autoencoder consists of two parts: an encoder and a decoder. The encoder is responsible for compressing the original data into a set of hidden state vectors, and the decoder is responsible for restoring this hidden state vector to the original data. In this process, the autoencoder attempts to minimize the reconstruction loss function, which is a measure of the difference between the original data and the reconstructed data. If a sample point has a large error after reconstruction, it indicates that it is an outlier.

[0074] In an optional embodiment, the method further includes:

[0075] receiving a query request sent by a user, wherein the query request carries information related to the supply chain of the industrial control system of the power generation group;

[0076] The supply chain related information is verified, and if the verification fails, a verification failure prompt message is returned to the user.

[0077] Specifically, in the embodiments of the present application, an autoencoder model can be used to detect anomalies in supply chain related data. Specifically, an autoencoder is an unsupervised learning algorithm that learns the intrinsic representation of data by compressing and decompressing the input data. The basic structure of an autoencoder consists of two parts: an encoder and a decoder. The encoder is responsible for compressing the original data into a set of hidden state vectors, and the decoder is responsible for restoring this hidden state vector to the original data. In this process, the autoencoder attempts to minimize the reconstruction loss function, which is a measure of the difference between the original data and the reconstructed data. If a sample point has a large error after reconstruction, it indicates that it is an outlier.

[0078] In summary, the embodiment of the present application discloses a supply chain security management method for the industrial control system of a power generation group based on a pruned autoencoder. The specific implementation scheme is: first, the supply chain related data of the industrial control system of the power generation group is obtained; then, the supply chain related data is detected using an autoencoder model, wherein the autoencoder model is a pruned autoencoder model. It can be seen that the embodiment of the present application uses an autoencoder model to perform anomaly detection on the supply chain related data, which can quickly discover abnormal situations in the supply chain and improve power supply reliability. At the same time, it can also reduce maintenance costs and improve operation and maintenance efficiency. In addition, the embodiment of the present application uses a pruned autoencoder model for anomaly detection, which can reduce the amount of data required for model training, shorten the model training time, and improve the model prediction accuracy.

[0079] Embodiment 2

[0080] The power generation group industrial control system supply chain security management system of the present invention includes:

[0081] An acquisition module, used to acquire supply chain related data of the industrial control system of the power generation group;

[0082] An anomaly detection module is used to perform anomaly detection on the supply chain related data using an autoencoder model, wherein the autoencoder model is a pruned autoencoder model.

[0083] The division of modules in the embodiments of the present application is schematic and is only a logical function division. There may be other division methods in actual implementation. In addition, each functional module in each embodiment of the present application may be integrated into a processor, or may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated modules may be implemented in the form of hardware or in the form of software functional modules.

[0084] Embodiment 3

[0085] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the supply chain security management method of the power generation group industrial control system are implemented, for example, including: obtaining supply chain related data of the power generation group industrial control system; using an autoencoder model to perform anomaly detection on the supply chain related data, wherein the autoencoder model is a pruned autoencoder model. The memory may include a memory, such as a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk memory, etc. The processor, the network interface, and the memory are interconnected through an internal bus, and the internal bus may be an industrial standard architecture bus, a peripheral component interconnection standard bus, an extended industrial standard architecture bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The memory is used to store programs. Specifically, the program may include a program code, and the program code includes computer operation instructions. The memory may include a memory and a non-volatile memory, and provide instructions and data to the processor.

[0086] Embodiment 4

[0087] A computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the steps of the supply chain security management method of the power generation group industrial control system are implemented, for example, including obtaining supply chain-related data of the power generation group industrial control system; using an autoencoder model to perform anomaly detection on the supply chain-related data, wherein the autoencoder model is a pruned autoencoder model. Specifically, the computer-readable storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. The volatile memory may include random access memory (RAM) and / or cache memory (cache), etc. The non-volatile memory may include a read-only memory (ROM), a hard disk, a flash memory, an optical disk, a magnetic disk, etc.

[0088] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0089] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0090] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0091] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0092] Those skilled in the art will readily appreciate other embodiments of the present invention after considering the specification and disclosure of the invention. This application is intended to cover any variations, uses or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary techniques in the art that are not disclosed by the present invention. The specification and examples are to be considered exemplary only, and the true scope and spirit of the present invention are indicated by the following claims.

[0093] It should be understood that the present invention is not limited to the exact construction that has been described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

[0094] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any way. Any simple modification, change and equivalent structural change made to the above embodiment based on the technical essence of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A method for supply chain security management of industrial control systems of power generation groups, characterized in that: The following steps are involved: S101: Acquire supply chain related data of the industrial control system of the power generation group; S102: Perform anomaly detection on the supply chain related data using an autoencoder model, wherein the autoencoder model is a pruned autoencoder model.

2. The power generation group industrial control system supply chain security management method according to claim 1 is characterized in that: The acquisition process of the autoencoder model is: S201: Obtain an initial autoencoder model, wherein the initial autoencoder model includes a plurality of neuron layers, and each neuron layer includes at least one neuron unit; S202: for each neuron layer in the initial autoencoder model, determine whether any neuron unit in the neuron layer satisfies the first condition, if not, execute step S203, if yes, execute step S204; S203: deleting the neuron unit in the neuron layer; S204: Adjusting the parameters corresponding to the neuron unit in the neuron layer.

3. The power generation group industrial control system supply chain security management method according to claim 2 is characterized in that: The first condition is any one or all of the second condition and the third condition, wherein the second condition is that the first parameter corresponding to the neuron unit is less than a first threshold, and the third condition is that the second parameter corresponding to the neuron unit is greater than a second threshold.

4. The power generation group industrial control system supply chain security management method according to claim 3 is characterized in that: The adjusting of the parameters corresponding to the neuron unit in the neuron layer comprises: When the first parameter corresponding to the neuron unit is less than a first threshold value and the second parameter corresponding to the neuron unit is not greater than a second threshold value, the first parameter corresponding to the neuron unit is set to zero.

5. The power generation group industrial control system supply chain security management method according to claim 4 is characterized in that: The initial autoencoder model also includes at least one skip connection, which is used to connect two non-continuous neuron layers in the initial autoencoder model.

6. The power generation group industrial control system supply chain security management method according to claim 5 is characterized in that: The process of anomaly detection is as follows: Reconstructing the supply chain related data using the autoencoder model to obtain reconstructed data; calculating a distance between the supply chain related data and the reconstructed data as a reconstruction error; When the reconstruction error is greater than a reconstruction error threshold, the supply chain related data is determined to be abnormal data.

7. The power generation group industrial control system supply chain security management method according to claim 6 is characterized in that: The anomaly detection also includes: Counting the reconstruction errors of all samples in the sample set to which the supply chain-related data belongs, and the average reconstruction errors of all samples in the sample set; When the average reconstruction errors of all samples in the sample set are greater than or equal to a sample average error threshold, the sample set is determined to be an abnormal sample set.

8. A power generation group industrial control system supply chain security management system, characterized in that: include: An acquisition module, used to acquire supply chain related data of the industrial control system of the power generation group; An anomaly detection module is used to perform anomaly detection on the supply chain related data using an autoencoder model, wherein the autoencoder model is a pruned autoencoder model.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the supply chain security management method of the industrial control system of a power generation group as described in any one of claims 1-7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by the processor, the steps of the supply chain security management method of the industrial control system of the power generation group as described in any one of claims 1-7 are implemented.

Images